Content
78%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The content is an unusually actionable operator reference: every tool, command, config field, and failure mode is concrete and executable, with safety rules and ambiguity checkpoints where operations are destructive. Its main weaknesses are inlined bulk reference material that belongs in separate files (no bundle files exist at all) and moderate repetition that costs tokens without adding guidance.
Suggestions
Split the ~25-row configuration table and the per-tool parameter reference into a references/ file (e.g. references/config.md and references/tools.md), keeping SKILL.md as an overview with well-signaled one-level-deep links.
Deduplicate the add_resource disabled-by-default guidance so it is stated once in the safety rules and merely cross-referenced from the Tool Selection table and the add_resource section.
Trim the five setup variant commands to the two most distinct patterns (peer-role and account/user-id tenancy) and mention the remaining flags in the config table instead.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense, tabular, and free of explanations of concepts Claude already knows, but it repeats the add_resource disabled-by-default rule in three places and lists five setup variant commands where two patterns would suffice, so a few tokens could be trimmed. | 4 / 5 |
Actionability | Fully executable guidance throughout: concrete commands ("openclaw openviking status --json", setup invocations with flags), complete parameter tables with defaults, real slash-command examples with arguments, and specific behavioral directives like "Try at least two concrete keyword variants before concluding archived detail is unavailable". | 5 / 5 |
Workflow Clarity | Clear intent-to-tool mapping, a status-first configuration flow with an explicit feedback branch ("Re-run `openclaw openviking setup ... --json`; branch on JSON `error`"), and a symptom/cause/first-action troubleshooting table; the destructive memory_forget flow has an explicit ambiguity checkpoint. Minor gaps (e.g., no explicit recovery loop after failed setup) keep it below 5. | 4 / 5 |
Progressive Disclosure | Section structure is good, but the ~25-row config table and the complete tool parameter reference are inlined in SKILL.md rather than split into reference files, and the closing "Reference Docs in This Repo" points to repository files (README.md, INSTALL.md, docs/...) that are not part of this skill bundle, leaving bulk reference content inline with weakly actionable navigation. | 3 / 5 |
Total | 16 / 20 Passed |